The disclosure relates generally to mechanical failure monitoring, and more particularly, to a system and turbine including a creep indicating member.
Mechanical part life, such as a rotor in a turbine, is dictated by one or more of several failure mechanisms. In turbine rotors subjected to high temperatures, creep and low cycle fatigue (LCF) are the prevalent failure mechanisms. Rotor failures can be catastrophic. A rotor burst can result in millions of dollars in damages and possibly loss of life. Consequently, rotors are designed for a useful life that is less than the predicted burst life, and is sufficiently less to greatly reduce the possibility of an in-service failure.
Many rotors have a limited creep life. Creep life prediction depends on many variables including temperature, stress, and material properties. Temperature and, through rotor speed, stress can be monitored during turbine operation. Material properties, however, vary from rotor to rotor. Unfortunately, the range of material properties can only be determined through destructive testing. Because of the variability in material properties, rotor lives, both predicted and actual, vary widely.
The extent of rotor creep can, for large rotors, be determined by measuring the rotor after a period of service. Typically, rotor diameter is measured, compared to the initial rotor diameter measurement, and correlated to a creep model to estimate the amount of creep, and hence the amount of life expended. Unfortunately, this approach requires good measurements of the new rotor, good data storage and retrieval, and disassembly of the turbine at the time of measurement. The disassembly requires expenditure of an extensive amount of time and costs.